[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 2/2] virtio-gpu/2d: add virtio gpu core code
From: |
Gerd Hoffmann |
Subject: |
Re: [Qemu-devel] [PATCH 2/2] virtio-gpu/2d: add virtio gpu core code |
Date: |
Mon, 08 Jun 2015 09:52:31 +0200 |
Hi,
> > +static uint32_t virtio_gpu_get_features(VirtIODevice *vdev, uint32_t
> > features)
> > +{
> > + return features;
> > +}
>
> Does this series rely on some other patches? Because in master,
> VirtioDeviceClass::get_features() is still uint64_t (*)(VirtioDevice *,
> uint64_t) (which results in "hw/display/virtio-gpu.c:886:23: error:
> assignment from incompatible pointer type" for me).
master was 32bit until recently (a few days back), now 64bit features
are finally merged and this mess should come to an end ;)
> (in an earlier series, this function was uint64_t ()(VirtIODevice *,
> uint64_t))
Earlier versions used to depend on unmerged virtio patches, this one was
for master ...
> > + if (ab->nr_entries > 1024) {
> > + qemu_log_mask(LOG_GUEST_ERROR,
> > + "%s: nr_entries is too big (%d > 1024)\n",
> > + __func__, ab->nr_entries);
> > + return -1;
> > + }
>
> Same question I had in said earlier series: Do you want to change this
> to 16384, because 1024 may be too small?
Oops, fixed.
> > +#define VIRTIO_GPU_MAX_RES 16
>
> Still unused. Still intentional? :-)
Fixed too.
cheers,
Gerd